Isa Speed Testing (Yak And 190 So Far)

[6./ZG26_Emil]

Just go on DED expert server or Wings of liberty. There is regularily when i am playing alone (thank god i have my mateys most of the time) when a YAK-1 sticks on my six and just keeps climbing after me up to to 9k height and i am unable to gian height in my suposedly superior performing(especially at those altitudes) F4.

Sorry but the high alt performance of the YAK-1 IS OP! There is no better word for that.

 

This is why the top speeds are less of an issue compared to climb rates, I've been testing the Yak a little and last night we did a 1v1 to test performance. I was flying the Yak-1 (I have very little flying time in it, under 1 hour) and at 6000m I could keep my Yak quite close to the 109 which couldn't spiral climb above me. Need to do more testing.

Edited March 6, 2015 by 6./JG5_Emil

[Trinkof]

This is why the top speeds are less of an issue compared to climb rates, I've been testing the Yak a little and last night we did a 1v1 to test performance. I was flying the Yak-1 (I have very little flying time in it, under 1 hour) and at 6000m I could keep my Yak quite close to the 109 which couldn't spiral climb above me. Need to do more testing.

Just a tip : to test it optimaly : 

 

Water rad 60% open

Oil rad 50% 

 

-> the less drag, with optimal temp around 100° for both temperature, this how it performs better, with nothing to change while in combat. 

[CaK_Rumcajs]

Every plane was different,some had brand new engines,some at the end of resurs,some repaired maybe twice or trice....This is fontline fighting,not NII VVS or Rechlin testing ground.

Even FWs deployed around Velikie Luki with I/JG51 had worn out engines.No spare to replace.You think they flew according Rechlin specs?

This sensation hunting gets boring,really.

My 2 cents

Look Brano. This is the very same we had been told in ROF. The Albatroses had worn out engines. The Camels were brand new and had the strongest engine version available. That's why the Camles hunted down Ablies like flies. And then a patch came out. It only pushed performance envelopes by couple of km/h in the right direction and people are satisfied. The DVa is faster by mere 7km/h now. And it makes the difference.

 

People who read threads like this one know all this. We know there are variations in performance. We know engines weare out. We know cold air has influence on aircraft performance. We know different Yak series had different performance. But we once were told we were given Yak 1 v. 69 so we expect it to perform in conjuction with that type. We do not expect the FW to be more worn out than the Russian planes. We do not expect the F4 to outperform the G2 by quite lot. We do not expect one particulr type to receive boost other types do enjoy. As a matter of fact of the two sides of the conflict one had been known to fly whatever was able to take off regardless of state. And we all know that it was the side to fly the Yak and La fighters. The 777 devs are not gods. They make mistakes as well as other humans. Their playground is limited by money, time, directives. People who are sceptical of what we are served to consume are in fact their best friends. They strive to make the game better. And they do it for free. They are not flawless either. I doubt the IAS - TAS conversions are calculated precisely. I more believe your findings measured with stopwatch on the map's grid. So it's nice you are bored by people disputing what we are given but face it, your boredom is not gonna stop questions when there is something strange.

Edited March 6, 2015 by II./JG1_Rumcajs

[Brano]

No need to write walls of text.The problem is vysotnyi korrektor in Yak.Only answer from devs can bring more light into this issue.Not multiple threads with the same content.No need to discover the wheel again.

[II/JG17_HerrMurf]

So just to understand :

 

Performance of the yak 69 in game is actually those of the 1B ?

 

So, as 1B was also over stalingrad, just regarding FM and not 3D models, the ingame confrontation we have is no fantasy , nor "broken FM", nor UFO plane ...

 

Problem is just 3D model is the wrong one ?

 

And another serious question : does the tests you ran about the IG yak are fitting with the same error margin as german plane, the real charts of the 1B ?

 

Because if the error margin is the same everyone could stop the "Bias" debate.

 

Thanks for your hard work ! O7

So, I claim no bias here on the part of the devs but I do have two problems with your statement. If we have a model 69 and the performance is for the 1B we have a problem. The type 69 should perform properly. If they want to introduce the 1B then port over the performance and create the 1B model.

 

Secondly, If there is a 4% positive error for one and a 4% negative error for the other then that is a significant problem. Even a 4% error seems pretty significant to me. In level flight thats what 17 kph (ish) difference. If both are carrying the 4% errors then we add them for a potential 34 kph (ish) difference. That is significant. Celestiale is suggesting there is already a 22 kph error (yet to be verified) for the Yak. That also is significant.

 

I'd like to see it much closer to 2% to be able to call the FM accurate.

 

If my maths are wr0ng, I'm OK with criticism.

[6./ZG26_5tuka]

Even FWs deployed around Velikie Luki with I/JG51 had worn out engines.No spare to replace.You think they flew according Rechlin specs?

My 2 cents

While this may be true it's probably worth mentioning that the devs already said in Alpha that all planes in BoS are modeled with factory fresh engines. That's why the Fw 190 should perform according to factory data so should any other plane ingame.

Edited March 6, 2015 by Stab/JG26_5tuka

[CaK_Rumcajs]

No need to write walls of text.The problem is vysotnyi korrektor in Yak.Only answer from devs can bring more light into this issue.Not multiple threads with the same content.No need to discover the wheel again.

Well this is the internet. We "still" enjoy freedom here. So there is no single individual who decides what is needed to discuss and what is not. Each of us is eligible to speak about whatever he/she likes as long as there is no interference with some regulations. Number of FM threads is not limited by any rules.

Also how can you be sure the problem is the vysotnyi korrektor? Has this been confirmed by the devs? What about the new finding about the FW overspeeding at 9000m? Isn't it related? Isn't this a general bug in BOS that planes overspeed above their max speed altitude?

[II/JG17_HerrMurf]

Why OP had to open yet another thread in General discussion,when there is allready topic in apropriate location?

 

http://forum.il2sturmovik.com/topic/14625-wrong-russian-fighters-performance-high-alts-again/

 

There was continuous improvement from series to series with Yaks.Or someone thinks that there were purely s69 present at Stalingrad in november 42?Early june 1942 production?Well,surviving 5 months of combat service....There were lots of series of Yaks flying around.Even some refurbished with PA engine (more of a rarity).And s99,which is first Yak1b (with high gargot) is October 42 production series.So only thing that devs need to do is rename the plane to s69-s85.

Every plane was different,some had brand new engines,some at the end of resurs,some repaired maybe twice or trice....This is fontline fighting,not NII VVS or Rechlin testing ground.

Even FWs deployed around Velikie Luki with I/JG51 had worn out engines.No spare to replace.You think they flew according Rechlin specs?

This sensation hunting gets boring,really.

My 2 cents

The problem with this is it has been stated several times by devs, really early in development, that all aircraft are factory fresh for consistency. It was also stated there is a 2% margin for error from flight to flight to account for production quality.

 

And (not you Brano) people tossing out such gems as, "well there were certainly more 1b's than Fw's around Stalingrad," in an FM thread is a complete red herring. I thought the testing as opposed to annecdotal information was refreshing for a change.

Edited March 6, 2015 by HerrMurf

[II./JG77_Manu*]

Why OP had to open yet another thread in General discussion,when there is allready topic in apropriate location?

 

http://forum.il2sturmovik.com/topic/14625-wrong-russian-fighters-performance-high-alts-again/

 

There was continuous improvement from series to series with Yaks.Or someone thinks that there were purely s69 present at Stalingrad in november 42?Early june 1942 production?Well,surviving 5 months of combat service....There were lots of series of Yaks flying around.Even some refurbished with PA engine (more of a rarity).And s99,which is first Yak1b (with high gargot) is October 42 production series.So only thing that devs need to do is rename the plane to s69-s85.

Every plane was different,some had brand new engines,some at the end of resurs,some repaired maybe twice or trice....This is fontline fighting,not NII VVS or Rechlin testing ground.

Even FWs deployed around Velikie Luki with I/JG51 had worn out engines.No spare to replace.You think they flew according Rechlin specs?

This sensation hunting gets boring,really.

My 2 cents

Read the topic post again, the whole one this time. Wrong Russian high alt performance is only a small part of it, so please spare us with your high horsing.

For month you have been attacking anyone who said, that the Yak is overperforming, and discredited his justification. Now that we have the clear data, and you are proven wrong, you come by, that we just should imagine that it's another plane that the Devs say? Please

[JtD]

Wrong Russian high alt performance is only a small part of it

Just like I said elsewhere, the Fw190 is overperforming above full throttle altitude just as much as the Yak is.

 

For month you have been attacking anyone who said, that the Yak is overperforming, and discredited his justification. Now that we have the clear data,...

It should be noted that Brano didn't attack everyone who said the Yak is overperforming or discredit his justification, and that he provided clear data before you did - so maybe you can for once stop your ad hominem attacks? It really doesn't add anything useful.

[II./JG77_Manu*]

Just like I said elsewhere, the Fw190 is overperforming above full throttle altitude just as much as the Yak is.

 

It should be noted that Brano didn't attack everyone who said the Yak is overperforming or discredit his justification, and that he provided clear data before you did - so maybe you can for once stop your ad hominem attacks? It really doesn't add anything useful.

Of course he did..the first time we were discussing about the Yak speed, he thought the given speed is with 2x100kg bombs attached...and "used" this as explanation, to "prove us wrong"...Just one example out of a lot..You don't know everything JtD, so don't talk like you would

[JtD]

He never attacked me or discredit my justification, and I said the Yak is overperforming. Maybe I just wasn't enough of an a***ole while saying so, but no, he didn't attack everybody.

 

And I don't see how having a different opinion equates to attacking people...

[GOAT-ACEOFACES]

It really doesn't add anything useful.

+1

[kendo]

Big developments...

 

http://forum.il2sturmovik.com/topic/15084-its-important-provide-data-developers-if-you-think-there-are/?p=240985

[6./ZG26_Emil]

Climbing from 6000-7000m I get 9.25 m/s. That seems very high for  Yak-1 doesn't it?

[303_Kwiatek]

Very high probably that too high. But these is also high probably that there would be a some people here which would claim opposite.

 

Objective people are very rare. Expecially here.

Edited March 6, 2015 by 303_Kwiatek

[6./ZG26_Emil]

Very high probably that too high. But these is also high probably that there would be a some people here which would claim opposite.

 

Objective people are very rare. Expecially here.

 

Dunno what book this came from or if I have the wrong end of the stick but I found this chart. Ignore the coloured lines as I assume they were used to demonstrate a similar issue in another game like WOWthunder

 

 

 

 

[Bearcat]

Very high probably that too high. But these is also high probably that there would be a some people here which would claim opposite.

 

Objective people are very rare. Expecially here.

 

 

Stop it please.... 

[GOAT-ACEOFACES]

Sadly, ROC testing is much harder than TSPA testing.. Unless there is an auto pilot setting to climb at a specific angle or speed.. Is there? Because without that, you really need a data logger to take out the spikes in the ROC data

[6./ZG26_Emil]

Sadly, ROC testing is much harder than TSPA testing.. Unless there is an auto pilot setting to climb at a specific angle or speed.. Is there? Because without that, you really need a data logger to take out the spikes in the ROC data

 

What spikes do you mean? If you enter a climb at the optimum speed and time it for 1000m I don't see how you can get higher rates, rather you might get lower ROCs. I understand why people are more focused on the maximum speeds as it's easier to log but in terms of simulating realistic air combat between the Yak and 109 (for example) the ROCs are just as important if not more. The 109 was a much better climber than the yakabove 3000-4000m and it should also be this way in game. Personally I don't mind that Yaks or LaGGs are a bit faster but the climb rates are vital.

 

Anyway unless this chart is either wrong or I misinterpret it (entirely possible) the Yak should have an optimum climb speed of around 7.5 m/s at 6000m dropping to 5.5/6 m/s at 7000m. If the chart is correct the Yak is climbing much faster than it should.

Edited March 6, 2015 by 6./JG5_Emil

[303_Kwiatek]

Stop it please.... 

 

Sorry you right i was wrong not expecially here but everywhere

What spikes do you mean? If you enter a climb at the optimum speed and time it for 1000m I don't see how you can get higher rates, rather you might get lower ROCs. I understand why people are more focused on the maximum speeds as it's easier to log but in terms of simulating realistic air combat between the Yak and 109 (for example) the ROCs are just as important if not more. The 109 was a much better climber than the yakabove 3000-4000m and it should also be this way in game. Personally I don't mind that Yaks or LaGGs are a bit faster but the climb rates are vital.

 

109 was better climber at all alts then all Russian planes from these peroid. I mean then Yak-1, La-5 not even mention Lagg-3.

Edited March 6, 2015 by 303_Kwiatek

[GOAT-ACEOFACES]

What spikes do you mean?

The spikes you get from not being able to maintain a constant climb speed.. It was easier for real pilots to do, because they could sense it, but in the sim we have very little feedback.. I know this to be true based on all the ROC testing we did in the original IL2, few if any could maintain a constant climb speed..

 

Every sim pilot that sent me his track file that I in turn extracted the climb speed (and other data) was surprised at how much the speed varied! And due to those 'spikes' you could get a very FALSE impression of the ROC.. It wasn't until..

 

Dang I forget his name? But some guy wrote an autopilot program that flew the plan via the DeviceLink interface.

 

But it want until than that we were able to obtain good ROC data.. Than shortly after that the ZINFOMOD came out where we could extract the ROC data directly from the game, which did away with the need to preform test flights anymore, other than for a sanity check of the ZINFOMOD.. JtD should remember those days, in that he was the one that turned me onto the autopilot program that many of use used for all sorts of testing.

 

But, we don't have that luxury anymore.. So, if your going to do an ROC test, I would highly recommend you make a track file of it and concentrate on maintaining a good climb speed.

 

Than, post test, re-watch the track file and write down your ROC values, and, as a sanity check, calc the ROC based of the time to climb values, to see that the average between two points (altitudes) matches what you think you got for an ROC value, smaller intervals of alt will make your average ROC value match the instantaneous value better.

[6./ZG26_Emil]

The spikes you get from not being able to maintain a constant climb speed.. It was easier for real pilots to do, because they could sense it, but in the sim we have very little feedback.. I know this to be true based on all the ROC testing we did in the original IL2, few if any could maintain a constant climb speed..

 

Every sim pilot that sent me his track file that I in turn extracted the climb speed (and other data) was surprised at how much the speed varied! And due to those 'spikes' you could get a very FALSE impression of the ROC.. It wasn't until..

 

Dang I forget his name? But some guy wrote an autopilot program that flew the plan via the DeviceLink interface.

 

But it want until than that we were able to obtain good ROC data.. Than shortly after that the ZINFOMOD came out where we could extract the ROC data directly from the game, which did away with the need to preform test flights anymore, other than for a sanity check of the ZINFOMOD.. JtD should remember those days, in that he was the one that turned me onto the autopilot program that many of use used for all sorts of testing.

 

But, we don't have that luxury anymore.. So, if your going to do an ROC test, I would highly recommend you make a track file of it and concentrate on maintaining a good climb speed.

 

Than, post test, re-watch the track file and write down your ROC values, and, as a sanity check, calc the ROC based of the time to climb values, to see that the average between two points (altitudes) matches what you think you got for an ROC value, smaller intervals of alt will make your average ROC value match the instantaneous value better.

 

Flying badly in a climb test will only result is worse rates of climb though. I'm not going to profess I know the Yak or fly perfect but I can get much higher rates than the graph shows.

 

And yep no problem I can do the latter points, I know the Yak was a good climber (in game) just from flying against it but I just was curious to see how good it is, I flew it about 10 times and got slightly varying figures but 9.25 m/s seems to be the average.

Edited March 6, 2015 by 6./JG5_Emil

[GOAT-ACEOFACES]

 

Oh, and you have to do the test several times! At different climb speeds!

 

Alot of WWII planes had a climb speed between 140 IAS and 170 IAS, so, you need to do one at 10mph intervals to find the 'sweeeet spot' wrt the best climb speed.

[6./ZG26_Emil]

Oh, and you have to do the test several times! At different climb speeds!

 

Alot of WWII planes had a climb speed between 140 IAS and 170 IAS, so, you need to do one at 10mph intervals to find the 'sweeeet spot' wrt the best climb speed.

 

Again Ace I'm getting too higher a figure so it doesn't matter. If I was struggling to get the right climb speed from the 109 then I would agree.

 

An autopilot program would be perfect for this I do agree.

[GOAT-ACEOFACES]

Flying badly in a climb test will only result is worse rates of climb though.

No necessarily..

 

You can climb steeply for short periods of time, but, you wont make it to 20kft doing it that way because you will burn the engine up. A constant climb speed is just another way of saying zero acceleration during the climb. So, you can get FALSE higher rates of ROC at some points and lower rates at others.

 

With that said, it should also be clear that you have to do the complete climb from sea level to max alt, anything in between and your not adhering to the definition of a ROC test.

Again Ace I'm getting too higher a figure so it doesn't matter. If I was struggling to get the right climb speed from the 109 then I would agree.

 

An autopilot program would be perfect for this I do agree.

Check your time to climbs, that is another way to tell if your getting false high values

[GOAT-ACEOFACES]

 

Mater of fact, I would be more interested in the time to climbs, in that has a way of taking into account the spikes in ROC due to non-constant climb speeds.. As long at the spikes are not too big.. Than you can calc the average ROC between two points in altitude, and if you make the intervals small enough, you will approach the instantaneous value, but not too small of a time interval in that you would than be dealing with the spike issue again. Based on my experience with the original IL2 ROC testing, I think 30sec intervals should be good.

[II./JG77_Manu*]

He never attacked me or discredit my justification, and I said the Yak is overperforming. Maybe I just wasn't enough of an a***ole while saying so, but no, he didn't attack everybody.

 

And I don't see how having a different opinion equates to attacking people...

You know, i wasn't such an "a***ole while saying so" from the beginning, i began to present my facts and Data in a normal way. But from the very beginning i got attacked again and again from the same sort of people(yes, i am not going to mince matters here, the SAME sort of people) who feared that their beloved Russian bird may be nerfed. They threw everything at me just do discredit me, and let people think it's right as it is, and "i am wrong".

OP is a special one, he pretty much jumps in every time, he thinks he can down me. It was the case in early Beta, when he threw in his false accusations in the first discussion about the 190. Then we were talking about Sabre-Mig, i stated something, he jumped in and tried to discredit me with his (false) scionism again.

And now instead of contributing something useful he drags this thread off topic with "was this really neccessary from op..." stuff. Only Starts useless discussions. Again and again.

Of course i am becoming a little pissed,i guess i proved him wrong one time to often and now he has an agenda against me. I am also only human, and I just can't stay as calm as Crump for example with this ongoing subliminal attacks.

Enough off topic

[JtD]

You need to take a chill pill and reading classes.

 

@Emil: Thanks for the climb info.

[CaK_Rumcajs]

Dang I forget his name? But some guy wrote an autopilot program that flew the plan via the DeviceLink interface.

If you are speaking about old IL2 then his nick was LesniHU. If you are speaking about Cliffs of Dover then his nick was Rumcajs.

[Crump]

 

 

i proved him wrong one time to often and now he has an agenda against me

 

That is common in this community. 

[GOAT-ACEOFACES]

If you are speaking about old IL2 then his nick was LesniHU.

Thats it!

 

If you are speaking about Cliffs of Dover then his nick was Rumcajs.

There is an autopilot for CoD? Never saw it, must have came out after I removed it from my HD

[6./ZG26_Emil]

Mater of fact, I would be more interested in the time to climbs, in that has a way of taking into account the spikes in ROC due to non-constant climb speeds.. As long at the spikes are not too big.. Than you can calc the average ROC between two points in altitude, and if you make the intervals small enough, you will approach the instantaneous value, but not too small of a time interval in that you would than be dealing with the spike issue again. Based on my experience with the original IL2 ROC testing, I think 30sec intervals should be good.

 

I'm not sure if I misunderstand but I timed the climb from 6000-7000m then divided by 1000m which gives the climb rate so we have both, if it's necessary we can give both figures when talking about climb rates.

[GOAT-ACEOFACES]

I'm not sure if I misunderstand but I timed the climb from 6000-7000m then divided by 1000m which gives the climb rate so we have both, if it's necessary we can give both figures when talking about climb rates.

If you divided distance by distance, that is not a RATE.

 

Rate of Climb, as the name implies is a RATE which in this case should be distance divided by time

 

average ROC = (Altitude Stop - Altitude Start) / (Time Stop - Time Start)

Edited March 7, 2015 by ACEOFACES

[CaK_Rumcajs]

There is an autopilot for CoD? Never saw it, must have came out after I removed it from my HD

It's not a publicly available software. It would be considered a cheat by many because of what it can do to optimize chosen flight regime and automate aircraft management. TF has the autopilot available for performance tests.

[6./ZG26_Emil]

If you divided distance by distance, that is not a RATE.

 

Rate of Climb, as the name implies is a RATE which in this case should be distance divided by time

 

average ROC = (Altitude Stop - Altitude Start) / (Time Stop - Time Start)

 

We have crossed wires. That is exactly what I did

 

Cheers

 

Emil

[GOAT-ACEOFACES]

We have crossed wires. That is exactly what I did

Ah good, because that is not what you said you did! Hence the crossed wires, but I figured it was a type-o or misspeak.. Just wanted to make sure

 

But as I noted earlier, starting an ROC mid alt, as you did, does not conform to the way ROC test were done, which was to start from the ground.. or at least as close to it as possible.

 

By doing it right way you eliminate the potential zoom climb (i.e. not meeting the zero accelerations requirement for ROC testing), and in light of the short run from 6000 to 7000 you could have introduced an accelerated zoom climb.

 

Hopefully all these little gotchas are starting to sink in as to why it is so important to make a track file of the test for review

 

It's not a publicly available software. It would be considered a cheat by many because of what it can do to optimize chosen flight regime and automate aircraft management. TF has the autopilot available for performance tests.

Yes, the C# interface on CoD, is something I did alot with to log data for testing, and the ability to write code to automate aircraft management is the same reason the orginal IL2 disabled allot if not all of the the DeviceLink interface when online.. Which sucked.. I always thought an 'e' meter is something that should be in flight sims, to give us feedback as to the planes energy state. Edited March 7, 2015 by ACEOFACES

[GOAT-ACEOFACES]

By doing it right way you eliminate the potential zoom climb

eliminate.. probably not the word I should have used, 'reduce the chance of' would be better, because a bad test pilot can still manage to induce accelerations that can screw up the results. In English, you have to maintain a nearly constant climb speed to keep the accelerations as close to zero as possible, which means the transition/changes in climb speed have to be done very gradually.

[6./ZG26_Emil]

eliminate.. probably not the word I should have used, 'reduce the chance of' would be better, because a bad test pilot can still manage to induce accelerations that can screw up the results. In English, you have to maintain a nearly constant climb speed to keep the accelerations as close to zero as possible, which means the transition/changes in climb speed have to be done very gradually.

 

I agree but I assume most people who attempt to do climb tests know the rules to stick by i.e. not starting your test at 400 Kph. By starting and ending at the optimum climb speed and doing your best to keep the speed constant should reduce the errors and at that speed there is little zoom climb available unless you try to cheat and pull the stick back at 6900m. Having done this test several times I think an autopilot program would get a fairly similar result maybe a little better.

[GOAT-ACEOFACES]

I assume most people

I use to assume that

 

But after processing dozens if not hundreds of original IL2 track files..

 

I realized my assumption was in error.